Back to Top
 
 
 
Combined Borehole Seismic and Electromagnetic Inversion for High-Resolution Petrophysical Assessment of Hydrocarbon Reservoirs
Project Number
DE-FC26-04NT15507
Goal

This project has two primary goals. The first is to advance the conceptual design of a new generation of deep-sensing borehole electromagnetic (EM) and seismic instruments that can provide 3-D images of reservoir flow units and of their petrophysical properties to distances of tens of meters from the wellbore. The second involves the development of efficient numerical algorithms, and computer codes for the joint inversion of borehole seismic and EM measurements. These algorithms will integrate data governed by multiphase fluid flow, visco-elastic, and electric phenomena into consistent 3-D images of lithology, porosity, permeability, and fluid saturation. Use of borehole seismic measurements in addition to EM measurements will reduce non-uniqueness and hence will improve the accuracy of the estimated petrophysical parameters. Deterministic and stochastic inversion techniques will be used to estimate reservoir-scale distributions of petrophysical properties.

Performer(s)

University of Texas at Austin, Austin, TX 
Lawrence Berkeley National Laboratory, Berkeley, CA

Background

Together with rock-core laboratory data, wireline and logging-while-drilling measurements remain the primary means of assessing in-situ properties of hydrocarbon-bearing rocks. However, because their length of investigation is less than 3 m, wireline logs can be biased by borehole damage and mud-filtrate invasion. New borehole measurements are needed to assess the spatial continuity and petrophysical properties of flow units to distances of tens of meters from the wellbore. Novel borehole measurements are also needed with 3-D capabilities to detect and assess reservoir flow units bypassed by wells. To date, no effort has been devoted to integrating borehole seismic and EM measurements into consistent 3-D spatial images of petrophysical properties.

Results 
To date, researchers have developed:

  • A 2.5-D time-domain finite-difference code to simulate borehole sonic measurements acquired in a vertical well in the presence of transversely isotropic axisymmetric distributions of elastic properties.
  • A 2.5-D time-domain finite-difference code to simulate borehole sonic measurements acquired in a vertical well in the presence of axisymmetric distributions of poro-elastic properties.
  • A 3-D time-domain finite-difference code to simulate borehole sonic measurements acquired in dipping wells in the presence of invaded rock formations exhibiting arbitrary elastic anisotropy.
  • A 1.5-D semi-analytical elastic code to simulate borehole sonic measurements acquired in vertical wells in the presence of radially varying rock formations exhibiting TI anisotropy.
  • A 1.5-D semi-analytical poro-elastic code to simulate borehole sonic measurements acquired in vertical wells in the presence of radially varying rock formations exhibiting TI anisotropy.
  • Developed an inversion code to estimate radial profiles of elastic properties from full-waveform borehole acoustic measurements acquired with monopole and dipole sources and multi-receiver acquisition.

Benefits 
The successful quantitative integration of borehole sonic and EM measurements will considerably improve current state-of-the-art technology to detect and assess the hydrocarbon potential of reservoir flow units undetectable with surface seismic methods. Results from this project will increase considerably the ability of oil and gas companies to detect and assess in-situ the petrophysical properties of rock formations penetrated by a borehole as well as their hydrocarbon potential. The inversion algorithms will be tested extensively and benchmarked against complex synthetic models constructed to replicate highly laminated Gulf of Mexico reservoirs, tight onshore reservoirs, dual-porosity carbonate systems, and naturally fractured formations.

Summary 
The goal of this project is to develop efficient deterministic and stochastic inversion algorithms to combine borehole sonic and electromagnetic measurements in the estimation of petrophysical properties of rock formations. There are three main developments considered in this project:

  • To develop efficient and accurate algorithms for the simulation of borehole EM and sonic measurements acquired in vertical and deviated wells.
  • To make use of these algorithms for the independent inversion of electrical resistivity (EM) and elastic properties (sonic) of rock formations.
  • To develop joint inversion algorithms that efficiently combine borehole sonic and EM measurements in the estimation of porosity, saturation, and permeability of rock formations. Both deterministic and stochastic inversion algorithms will be considered to combine the two types of measurements via empirical correlations among petrophysical, electrical, and elastic properties (e.g., Archie and Biott-Gassmann). These correlations will consider the effect of mud-filtrate invasion on sonic and electromagnetic measurements acquired in boreholes.

Thus far, researchers have accomplished the first and are currently working on the second item. The team from LBNL is working on the stochastic formulation of the combined borehole sonic-electromagnetic inversion.

Researchers currently are developing an inversion code to estimate porosity and radial profiles of saturation and elastic properties jointly from EM and full-waveform borehole acoustic measurements.

Current Status

(February 2008) 
Researchers have developed and successfully tested a new method to estimate dry-rock elastic properties of rock formations in situ. The method combines the joint numerical simulation and inversion of borehole resistivity and sonic measurements. They have tested the joint inversion method on both synthetic and field data sets assuming radial variations of fluid distributions resulting from the process of mud-filtrate invasion.

The project is complete. The final report is listed below under "Additional Information".

Funding 
This project was selected in response to DOE’s 2004 Oil Exploration and Production solicitation DE-PS26-04NT15450, Advanced Diagnostics and Imaging.

Project Start
Project End
DOE Contribution

$799,238 

Performer Contribution

$199,925

Contact Information

NETL – William Fincham (william.fincham@netl.doe.gov or 304-285-4268)
U.T.-Austin - Carlos Torres-Verdin (cverdin@uts.cc.utexas.edu or 512-471-4216)

Additional Information

Final Report [PDF-4.43MB]

Publication
The first semi-annual report was submitted in June 2005. The second semi-annual report was submitted in January 2006.

Semi annual report was submitted in January, 2007

Figure 1. Input sonic waveforms numerically simulated, assuming radial variations of elastic properties away from the well, a monopole source—a Ricker wavelet centered at 3 KHz—and 33 acoustic receivers.
Figure 1. Input sonic waveforms numerically simulated, assuming radial variations of elastic properties away from the well, a monopole source—a Ricker wavelet centered at 3 KHz—and 33 acoustic receivers.
Figure 2. Comparison between actual and estimated radial profiles of density, P-wave velocity, and S-wave velocity for three levels of synthetic Gaussian noise included in the input sonic waveforms shown in Figure 1.
Figure 2. Comparison between actual and estimated radial profiles of density, P-wave velocity, and S-wave velocity for three levels of synthetic Gaussian noise included in the input sonic waveforms shown in Figure 1.
Probability density functions (PDFs) of estimated seismic P-wave and S-wave velocities (Km/s) , and bulk density (g/cm3) for the two types of rock assumed in the inversion. Black lines identify the estimated PDFs and blues lines identify the true values.
Probability density functions (PDFs) of estimated seismic P-wave and S-wave velocities (Km/s) , and bulk density (g/cm3) for the two types of rock assumed in the inversion. Black lines identify the estimated PDFs and blues lines identify the true values.
Description of a dipping and anisotropic layered formation model assumed in the numerical simulations of borehole sonic measurements.
Description of a dipping and anisotropic layered formation model assumed in the numerical simulations of borehole sonic measurements.
Comparison of the simulated x-directed velocity for different dip angles of the anisotropic layered formation model shown in the previous illustration.
Comparison of the simulated x-directed velocity for different dip angles of the anisotropic layered formation model shown in the previous illustration.